Upload
lethuy
View
216
Download
0
Embed Size (px)
Citation preview
..
~
I
PAP 149 1
TRACTIVE FORCE FLUCTUATIONS
AROUND AN OPEN CHANNEL
PERIMETER AS DETERMINED FROM
POINT VELOCITY MEASUREMENTS
by
Phillip "F." Enger
A paper to be presented at the ASCE Convention Phoenix, Arizona, April 10-14, 1961
..
"
TRACTIVE FORCE FLUCTUATIONS AROUND AN OPEN CHANNEL PERIMETER AS DETERMINED
FROM POINT VELOCITY MEASUREMENTS
by
Phillip "F. 11 Enger*
SYNOPSIS
The study described was conducted primarily to investigate the
possibility of using point velocities occurring in an open channel
to determine the tractive force (boundary shear) distribution on
the perimeter of the channel.
In a straight trapezoidal channel constructed with a well-graded
sand-gravel boundary, various discharges were established and
vertical velocity traverses were obtained. Velocities were obtained
by use of Pitot banks, and were plotted and shown to follow a loga
rithmic formula. Using von Karman•s logarithmic velocity distribution
law, the boundary shear distribution was calculated from the velocity
distribution. The average of this boundary shear was determined and
compared with the average obtained from du Boys formula. General
agreement of the two averages is indicated at all dischargeso
The mathematical method of least squares was used to fit a quadratic
tractive force distribution curve to the data. The data indicate
boundary shears distributed around the perimeter of the channel tested
*Hydraulic Engineer, Hydraulic Laboratory, Bureau of Reclamation, Denver Federal Center, Denver, Colorado.
,,
fluctuated considerably. A deviation of the tractive force fluctu
ations from the average curve was calculated.
INTRODUCTION
Although the practice of excavating channels in earth and conducting
water through them goes beyond the beginning of recorded history,
there are some phases of this science which are not fully understood.
One of these includes the distribution of forces acting on the
channel boundary due to the flowing water, and stability of the
boundary as a result of these forces.
Fluid forces acting on a channel boundary are termed tractive forces.
Knowledge of their distribution and magnitude around a channel perim
eter is important if' a stable design through a given material is
desired.
In the past, many investigators have been concerned with developing
equations tor velocity distributions in turbulent flow. This study
was made to investigate the possibility of using the resultant
formula to determine the tractive force distribution on the perimeter
of an open channel.
For this study, a trapezoidal-shaped channel of a well graded sand.
gravel mixture was formed, and point velocity traverses were obtained
2
for several discharges. Computations of boundary shear conditions
were made from these point velocities.
SHORT BACKGROOND AND APPROACH TO THE PROBLEM
In the late 19th Century, Osborne Reynolds expressed the apparent
shearing stress between parallel planes in the flow due to transfer
of momentum as :
where
~•shearing stress between planes
f • density of the fluid (neglecting variations)
u' = velocity fluctuation in direction of flow
v-' a velocity fluctuation perpendicular to direction of flow
The bar indicates mean values, and the negative sign is
used because of the general association of positive v''
with negative u' and negative 11"' with positive u'.
These turbulent stresses are so large in comparison to laminar
friction that laminar friction is usually neglected in turbulent
flow, except in the "laminar sublayer" where laminar friction is
predominant.
Prandtl (l)* introduced the idea of a "mean free path" or "mixing
length," .,R. , defined as the average distance, perpendicular to the
*Numbers refer to bibliography.
3
•
main flow, traveled by a particle before it accommodates itself' to
new surroundings. Prandtl assumed u ' proportiona.l to ,l ;, the
mixing length and the velocity gradient, and ! v' \ proportional to
I V: I • He then wrote the equation for the shear stress, by simple
substitution as:
In the preceding assumption and following asBW11ptions,f is necessarily
assumed to be small.
To determine the mixing length, von Karman (2) assumed that the
turbulent exchange was independent of the viscosity and the local
flow pattern was statistically similar at every point with only time
and length scales varying. Using these assumptions, he set the
mixing length proportional to:
or
Substituting into the shear stress tormu1a results in: l&u...V
?;: - K2.f L&;)
or
~j - (~)'
= ~~~ ~~a
4
Replacing 7;; by its value at the boundary, ?;' , and finding one
solution for the differential equation results in:
where
f • a constant having dimension of a length, is small in
comparison with y and is neglected.
d • a constant having dimensions of a length.
The preceding equation written in the form of:
--« = -' rz: k 2-K J_T d
is considered the "universal velocity distribution near the wall,"
and is developed as a function of the distance from the wall as:
hl :: + ff h:; + /"fJ) where f(y) depends on the fluid viscosity or the roughness of the
wall.
Von Karmen determined the best value of k to be o.4, and Keulegan (3)
developed the equation for open channels with smooth walls as:
:;-* = s.s + s.75 ~ (j ~u*)
and for rough walls as:
t; = [3.5 +-5.7S~ (-i) where:
ks .. a roughness factor.
5
Einstein (4) combined Keulegan's formulas into one general formula
covering smooth and rough walls by writing them in the form:
where:
~ = S".75 ~ ( 30·~,/f ) ~ 5. 7S ~ (~
:X = f ( iV a corrective parameter
..6 .:= -i-s .x
Using this general equation in the form:
~= c4(¥) and writing the equation for two points near the bed and in the range
where the equation expresses the velocity distribution results in:
where:
= L ~ cc~:~) C ~ cc~~·)
Subtracting the velocity at y1 from that at y2
u - (,/ = I J ~ C [ /_ ~... - J' ...... ~I J ,,. !f vv,r --::1 ~ ---J &>
and writing the last term as
log C1 + log y2 - log A - log c1 - log y1 + log A
and canceling, results in
log y2 - log Y1 = log ( ~~)
6
The eqµation now assumes the :form:
u.'j .. -u~, =- u~ C ~ (f.) Solving for 7;
This eqµation can be simplified if y2 and y1 are used as constants
giving:
where:
I:f y2 and y1 are chosen to result in c2 = l the eqµation is further
simplified.
This eqµation indicates that boundary shear at the bed, commonly
called tractive :force, varies with the sqµare of the velocity gradient,
and that by choosing two points from the velocity gradient the shear
at the boundary can be determined.
The eqµation is, of course, limited to the assumptions made in the
basic derivations.
This equation was used to establish the tractive force distribution
around the perimeter of the trapezoidal channel. To check the results,
7
the average tractive force was also obtained from the du Boys formula,
which can be derived as follows:
If water is flowing through a channel with a constant cross section,
A1 the total weight of the water volume over a length Lis:
W = A oL
where: o a unit weight of water
The force acting normal to the channel slope is W = A o L cos ~ . where : Tan ex. is the slope of' the channel bottom. The total force
acting along the slope is:
F • o AL sine:.,(_
This force may be expressed as a force per unit area by dividing by
·the wetted perimeter, P, and the length over which it acts, L. This
will be recognized as the sverage boundary shear., or tractive force 'J;:. •
J;: • ~A sin O!...
Noting A is equal to the hydraulic radius, and that for small angles p
sin~ is approximately equal to Tan °'" , or the canal slope, the
formula results in: x= oR. s
For wide rivers R is approximately equal to the depth, d, and for
nonuniform flow the slope of the energy gradient is used in the
formula.
8
TEST ~IFMENT
A laboratory test flume 72 feet long, 8 feet wide, and 24 inches
deep, Figures 1, 2, and 3, was used for the study. TW'o to one
side slopes made of wooden framework covered with a concrete mix
were built into the flume. Over the side slopes and on the channel
bottom, a sand-gravel mixture, analysis curve shown in Figure 4, was
placed. The channel section was formed by use of a template as shown
in the photograph of Figure 5. The template was controlled by guide
strips with a slope along the channel of 0.00545 foot per foot. Four
static disks connected to stilling wells, Figure 6, were installed on
the center line of the channel at Stations 0+10, 0+30, 0+50, and 0+70
(Station 0+00 was near the channel entrance).
Velocity readings were obtained with a Pitot tube and cylinder bank,
Figures 7 and 8a. The Pitot banks were designed to allow their
location to be readily changed. A photograph of the Pi tot banks
mounted on the channel is shown in Figure 8b. The Pi tot banks were
connected by means of rubber or plastic tubing to a manometer board
which was adjustable for various slopes, Figure 8c, and on the top
Pitot tube were located taps for measuring static pressure for
checking the static disk readings.
A standard mixture of aerosol and dye was used for bleeding the
Pi tot banks. Tu.be calibration was checked w1 th the standard mixture
9
for the slope set on the manometer board, and Pitot tubes and
cylinders were bled in a reverse direction to remove all air from
the system.
A 12-inch laboratory pump was used to pump water from a laboratory
sump through the model, and water discharge was measured by Venturi
meters to check the discharge determined from the velocity readings.
OPERATION OF THE MODEL
The general procedure used in conducting a test was as follows:
1. The model was slowly backfilled.
2. The desired discharge was slowly set.
3. While the discharge was being set, a near constant water
surface elevation was maintained at Station o+66.
4. The model was operated continuously for an approximate
24-hour period.
5. While the model was being operated, point velocities were
determined by means of the Pitot banks. Velocities were taken
throughout a vertical half section of the channel at Station o+66,
and .as near the boundary as possible. Velocity profiles were
taken at 4-inch intervals across the right side of the channel
(looking downstream) and the tubes and cylinders were frequently
interchanged to check results.
6. At frequent intervals, the hook gage elevations in the wells
connected to the static disks were ottained and the discharge was
checked.
10
7. After 24 hours, the model was turned off and slowly drained.
8. After draining, the material which had deposited in the tail
box was collected and weighed. It was then mixed and a sample
was taken for analysis.
9. Cross sections were obtained at Stations o+lO, o+30, o+50,
and o+70.
10. Point samples of' the sand-gravel mixture around the perimeter
and on the channel center line were taken and analyzed.
11. Observations were made during and ai'ter each run and photo
graphs were obtained.
12. Samples were replaced for the next test.
A discharge of 3.70 cubic feet per second was set for the first test.
The discharge was gradually increased for each additional test. Eight
tests were conducted, and the final discharge was 8.25 cubic feet per
second.
DATA OBTAINED
Average discharges are shown in Table 1, and relative water surface
elevations, obtained from hook gage readings, are shown in Table 2.
Figure 9 shows channel cross sections, which gradually developed
throughout the tests, and pertinent test data are listed in Table 3.
Point velocities obtained from Pi.tot readings may be obtained from
reference (5), and a typical series of velocity profiles are shown
plotted in Figure 10.
11
ANALYSIS OF DATA
Average Tractive Force
Before and after each test, cross sections were obtained at
Stations 0+10, 0+30, 0+50, and 0+70, and during tests a number of
water surface elevations were obtained at each of these stations.
Following tests, average channel cross sections were plotted and the
area of flowing water (A), wetted perimeter (P), hydraulic radius (R),
average velocity at the station (Va), and the velocity head (hy) were
determined.
r v.;:- n 2.
Writing Manning's equation in the form 2. J Se = L /. ~ 9' R.. 'fJ
the slope of the energy gradient, Se, is indicated as a :function of
the roughness, n. As Va and R were known, a value of n was assumed
and Se calculated at each station. Using the average of the Se values
at any two stations for the average energy slope occurring between
these stations, a water surface profile was calculated (by starting
at any given point) and compared with the measured water surface.
Using then value which resulted in a calculated water surface which
most closely fitted the measured water surface, the Se values at
Stations 0+10, o+30, 0+50, and 0+70 were determined. Average tractive
forces, X, were then calcuJa ted from the formula:
7::= o R Se
12
Curves were drawn through the points, resulting in the graph,
Figure 11, of the average tractive force acting along the channel.
Velocity Profiles from Point Velocities
During each test, velocities were obtained by the Pitot tube and
cylinder banks, Figure 7. When obtaining velocities, the banks
were set in the desired location and allowed approximately 5 minutes
to stabilize. Approximately four manometer readings were obtained at
each position at 5-minute intervals. If manometer readings indicated
a continued change during this time, additional readings were obtained
until stability occurred. Numerous velocity readings were obtained
in the lower portion of the flow near the boundary (below o.4 depth),
and several readings to help establish the velocity curve were obtained
above o.4 depth.
Depths for plotting velocities were determined from the position of
the boundary at the time velocity readings were started at a given
vertical, and velocities near the boundary were determined first.
On side slopes, the velocity distribution normal to the boundary
was plotted.
For all tests, the manometer board was sloped to increase the dis-
placement created by the velocity head by a factor of 20, or:
d - 'JQ L :::. 2.0 v,.' - C, 11.,r 2-g
u;=t.79'1/;T
13
where:
d = displacement on sloping manometer board
hy = velocity head
V~ = average velocity at the point
g = acceleration of gravity (assumed to be 32.2 feet
per second per second)
Velocity plots for all tests were made in the form of y/d versus the
velocity, Figure 10, on semilog graph paper. Straight lines were
found to generally fit the velocity points.
Tractive Forces Determined from Point Velocities
The equation
where:
was used to determine the boundary shear from the point velocities.
To determine the boundary shear at a point where a vertical velocity
distribution had been determined, points from the velocity curves
were read so that
:::. I. 7f" 6 ff.
_::---
d
Using the square of the difference of the two velocities, the shear
acting at the boundary was determined.
14
Average tractive force distributions are shown plotted in relation
to the channel boundary in Figures 12 and 13, and plotted in a
dimensionless form for comparison in Figure 14.
For the points shown on Figure 14, a quadratic equation of best fit
was determined to be
Jr = 1.27 + 1.05 Dr - 2.31 Dr2
where:
'];=ratio of tractive force acting at a boundary point
divided by the average tractive force.
Dr= distance of the boundary point from the channel
center line divided by total distance from
center line to water surface edge.
As shown by Figure 14, the distribution was erratic for each test,
but did show a general trend. Because of the general trend, a
quadratic equation was fitted to the data by the least squares
method.
Inspection of Figure 14 shows a maximum condition in the best fit
curve occurring to the right of the center line. This may have been
due to secondary currents. Its position may readily be determined
from the derivative, as Dr= 0.227, at which point J; = 1.38. This
would indicate that a tractive force 1.38 above the average generally
occurred in the channel.
15
The standard deviation for the data plotted was determined to be 0.28,
indicating that in the channel tested tractive forces, as determined
by this method from the velocity distributions, usually fell in the
range .. D.a T :: /. 2 7 f- !. 0 SD.,,. - 2. 3 I r - - - - - ± O. 2 B r
where units are as previously defined.
Data indicate that, generally, the average maximum tractive force
occurring on the channel bottom as determined by this method in this
channel was approximately 38 percent greater than rnse, and as the
side slope started at approximately half the distance to the water
surface edge, that average values approximately 22 percent greater
than oRSe occurred on the side slope. These forces are larger than
those usually thought of as maximum tractive forces occurring on the
boundary.
Comparison of Average Tractive Force Determined by Ja. = oRSe and Average Tractive Force Determined from Velocity Distribution
Assuming a linear distribution of tractive forces between points at
which the boundary shear was determined, an average tractive force
from point velocity data were determined from ..-l'l )I,
2:= * T .6.L.._ 12-
L
where:
Ti = the average tractive force occurring over interval L:::. Li
L = total length of wetted perimeter over which velocities
were measured
n a number of intervals
16
The average tractive force obtained by this method is given in
Table 1, as is the average ·tractive force obtained from 7;,, • 0R5e.
Figure 15a shows a plot of the two average tractive forces. In
general, there appears to be no consistent trend for tractive forces
obtained from point velocities to be either higher or lower than
those obtained by the conventional formula. The tractive forces
obtained from the point velocities are more erratic than those
obtained from the conventional. formula. This is understandable
because of the sensitivity of the tractive force to the velocity
gradient. In general, a slight change in the velocity gradient will
result in a considerable change in the magnitude of the tractive
force. If insufficient velocity points are determined, the tractive
force obtained from this method would be of questionable value.
Che.pge in Roughness of Channel
As the tests progressed, the roughness of the channel increased. The
increase in roughness was probably due in part to the finer material
being transported from the surface and into the tail box, while
larger material stayed in place . Figure 15b shows the roughness
increase plot t ed in terms of Manning• s n value. As may be noticed,
the value i ncreased from 0.0144 to o.018o.
DISCUSSION
Plots of point velocity data i ndicated the vertical velocity dis
tribution closely f ollowed a logarithmic form. Secondary currents
17
were neglected in the study, and it is assumed they had little
effect on the results.
A large source of error in determining the tractive force distribu
tions from point veloci~ies can come from establishing the slope of
the velocity gradient, as a slight error in slope will result in a
considerable difference in the computed tractive force. For example,
when the velocity gradient is near 70° from the horizontal, a
4 percent error in setting the slope would result in approximately
30 percent error when calculating the tractive force. Care should
be taken in selecting a line through the velocity points, and numerous
and accurate velocity points should be obtained when the velocity
gradient is to be found by this method. Care must also be taken in
establishing the point and the boundary, as f0-p, a given point near
the boundary a small error in establishing the location of the
boundary or point will result in a large error in the y/d value.
For this reason, when operating with a movable bed, it is not always
best to use the lowest points in establishing the velocity gradient.
Points below o.4 of the depths should be used , but if the curve has
considerable deviation near the bottom-most points, the magnitude of
the possible error should be checked.
Many attempts have been made to determine the distribu.tion of
tractive f orces in an open channel. Lane ( 6) found the maximum
18
tractive force on the side slope to be equal to approximately
0. 78 o dSe, and that on the bottom to approach f"dSe. Other
investigators have determined the distribution of tractive forces
varies with the side slope of the channel, but is unaffected by
the size of the section.
CONCWSIONS
On the basis of these tests and the analysis used in this report,
it is concluded:
L The velocity distribution in the open channel tested
followed a logarithmic equation over the entire cross section.
2. Secondary currents of the magn1 tude encountered in the
study did not appreciably affect the logarithmic equation.
3. Boundary shears, determined from point velocities, approached
maximum values near the center of the channel, and were approxi
mately 40 percent greater than the average tractive force as
obtained from the formula, 'J;, =ORSe•
4. Boundary shears, determined from point velocities, became
less on the side slope, but near the toe of the slope still had
average values greater than those obtained from');. =ORS8 •
5. Boundary shears, determined from point velocities approached
small values near the water surface edge.
6. Boundary shears distributed around the perimeter of the
channel tested fluctuated considerably, with a standard deviation
of approximately 28 percent from the average point value.
19
7. Distribution of average boundary shear across the channel
cross section (determined from fitting a quadratic equation to
the scatter diagram by the least squares method) followed the
formula
Jr • 1.27 + 1.05 Dr - 2.31 Dr 2
where Jr is a tractive force ~atio and Dr is a distance ratio.
8. When computing boundary shears from point velocities, it is
important to use numerous and accurate point velocities occurring
below o.4 depths. Location of the boundary and each velocity
point should be accurately determined.
9. Average boundary shears computed from point velocities
generally agreed with those computed from du Boys formula.
20
REFERENCES
1. Prand tl, F. , "Uber die ausge bildete Turbulenz," Proceedings of Second International Congress for Applied Mechanics, ZUrich, 1926.
2. von Karman, Th., "Goettinger Nachrichten," Proceedings of Third International Congress for Applied Mechanics, Stockholm, 1930.
3. Keulega.t''I, G. H . , "Laws of Turbulent Flow in Open Channels," Journal of Research of the National Bureau of Standards, u. s. Department of Commerce, National Bureau of Standards, Research Paper RP1151, Volume 21, December 1938.
4. Einstein, H. A., "The Bed-Load Function for Sediment Transportation in Open Channel Flows," Technical Bulletin No. 1026, u. s. Department of Agriculture, September 1950.
5. Enger, P. ''F .;' "Tractive Force Distribution around the Perimeter of an Open Channel by Point Velocity Measurements," Master's Thesis, submitted to the University of Colorado, August 1960.
6. Lane, E. W., Progress Report on Results of Studies on Design o-f Stable Channels, Hydraulic Laboratory Report No. HYD-352, u. s. Department o-f the Interior, Bureau of Reclamation, June 1952.
21
TABLE I
DISCHARGES AND '.ffiACTIVE FORCES COMPARED
Test No.:Discharge: Discharge: Average : Average : Average read on :calculated:tractive :tractive :Manning's Venturi from force force : "n" meter velocity : from : from : value
cfs reading : ~ 'o!eSe : 'FC{tly;/4:f: : cfs :at station:at station.
: : :0+66 :0+66; • · · lb/f't2 • lb/f't2 . . . . -2 . -2 ·
•••••••• i •..•.•••• i ••••••.••• i.~.,lQ •.•• i.7.llQ •••• , •••••••••
1
2
3
4
5
6
'7 I
8
3.70
4.72
5.51
6.03
6.62
7.16
7.70
8.25
3.64
4.60
5.6o
6.05
6.77
7.29
7.66
7.85
0.65
1.00
3.02
4.33
5.27
o.84
1.16
1.22
2.09
3.12
3.39
4.03
4.90
0.0144
0.0144.
0.0144
0.0150
0.0155
0.0155
0.0167
.: 0.0180
TABLE 2
RELATIVE WATER SURFACE ELEVATIONS
Test No. . Station . Station . Station: Station . . . : 0+10 : O+JO : 0+50 : 0+72 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •
1 . 5 .8381 5 .8373 : 5.8424 5.8415 . . . . . . . . .
2 5.7628: 5.8515 5.8528: 5.8574 : :
3 5.8660 : 5.8534 5.8544 : 5.8570 : . : .
4 5.8678 : 5.8438 : 5.8440 : 5.8387 : . : .
5 5.8746 : 5.8508: 5.8425 : 5.8367 : :
6 5.9031 : 5.8770 : 5.8680 : 5.8572 : . .
7 5.8860 5.8498: 5.8309 : 5.8118 : :
8 5.9134 5.8840 . 5.8589 : 5.8338 .
Table 3
:Hyd.raul.lc:Average : : wei;i;ea. Test:Station:perimeter,:Area: radius, :velocity,:V2
X 102 : : feet : f't2 : feet ft/sec :2g . . . . 1 : 0+10 5.46 :2.3l: o.423 1.6o 3.97
: 0+30 5.73 :2.79: o.487 1.33 . 2.74 . : 0+50 6.17 :3.32: 9.538 l.ll l.91 : 0+72 6.73 : 4.06: o.6o3 0.911 l.29 . : .
2 : 0+10 5.4o :2.16: o.4oo 2.19 7.24 : 0+30 5.86 :2.85: o.487 1.65 4.25 : 0+50 6.21 :3.41: 0.549 1.38 2.97 : 0+72 . 6.82 :4.19: 0.614 1.13 1.98 . . . . .
3 : 0+10 5.48 :2.44: o.445 2.26 7.95 : 0+30 5.98 :3.02: 0.505 1.82 5.17 : 0+50 6.41 :3-56: 0.555 1.55 3.72 : 0+72 . 6.88 : 4.20: 0.610 l.31 2.68 .
: . . 4 : 0+10 5.92 :2.58: o.436 2.33 8.46
: 0+30 6.03 :2.91: o.482 2.07 6.67 : 0+50 6.47 :3-39: 0.524 l.78 4.92 : 0+72 6. 72 :3,90: 0.581 1.54 3.71 . . . . . .
5 : 0+10 5.68 :2.80: o.493 2.36 8.66 : 0+30 6.06 :3.20: 0.528 2.07 6.63 : 0+50 6.28 :3.38: 0.538 1.96 5.95 : 0+72 6.98 :3.94: 0.564 1.68 4.52 . .
6 : 0+10 5.75 :2.73: o.475 2.62 10.7 : 0+30 6.22 :3.23: 0.519 2.22 7.6o : 0+50 6.46 :3.49: o.54o 2.05 6.6o : 0+72 7.00 : 3.86: 0.551 1.86 5.4o . . . . . .
7 : 0+10 6.06 : 3.04: 0.501 2.53 10.0 : 0+30 6.17 :3.21: 0.520 2.4o 8.90 : 0+50 6.51 :3.53: 0.542 2.18 7.4o : 0+72 6.76 :3.66: 0.542 2.10 6.Bo
8 : 0+10 6.22 :3.05: o.490 2.70 11.3 : 0+30 6.33 :3.38: 0.533 2.44 9.25 : 0+50 6.62 :3.67: 0.555 2.25 7.83 : 0+72 6.96 :3.83: <ih550 2.15 7.19
-~lZf ---1---------+---=+---''---------------l-------------'-----------f+----+--l
_j
Q 0 r<)
+ + 0 0
~ ~ en en
y \ I y
t r r r i r y r I I I : ' : I
FLOW ' 0
' _-_,.. ~-----cl,--- - -',
0 0 ,0
--·i e--
+ + 0 0 ~:\ ~ TAIL GATE------- ~ en en
r I r r r r I r ~ i - l / ,.,
~),
.. ~?TI=)=' =======o:;;::==::::::::!::::==============o.:============j====;o============:::::::1+====;0;::::=1:=======:j a '
I
l n l I
) l J ) J ( J \ ) ) :
TAIL Box·· ,1
'._ HEAD BOX -< ------------------ -- ---\- - --------- - - - - --- ------ - - ----72'-o" -- ---- --- --- - ----------- ------------ - ------ ~ -- - --- -- - -- -- -- ~ \ \ ~ I
\ PLAN L i ,!, 0 5 10 20;,,-- ~
STATIC DISKS WITH STILLING WELLS·') SCALE OF FEET A ,'' en
1 CALIBRATED MANOMETER BOARD WITH ADJUSTABLE,/ /_,// SLOPE FOR PI TOT BANK READINGS ---------- -<.,,\
SECTION A - A 0
SCALE OF FEET
S=0.00545
ELEVATION
GENERAL PLAN OF FLUME USED FOR TRACTIVE FORCE STUDY
.,, G> C ;u ,,,
Figure 2
FLUME PREJ;>ARED FOR TEST 1
MODEL IN OPERATION DURING TEST 5
SIEVE ANALYSIS
•200 •ioo •so •4-0 •30 •16 •t0 Its .4 i US STANDARD SERIES I
:
: , : ,
I , -,
I , I 7
7 I
I
I , '
, ; , ,
7 7
! , . " • /
I , I , ,,
- ·· , I , I , I , I , I ,
I
I I r, I ,, I
~ I I , I I , I T - ' - ! :
<Deno ~ r ,ti on C.O ,-... cm O'JO N ~ "' W") co,-..., a,,o q q ~ ·1 . 1· - I 1-0.42 2.0
.074 .14-9 .297 .590 1.19 2.38 4.76 9.52 DIAM ETER OF PARTICLE IN MILLIMETERS
CLEAR SQUARE OPENINGS 3• ,r 3• --------
0 0 ~ ~:ii~ :i:i:~ N ~
19.1 38. 1 76.2
,; 6"
I I I
;
I 127 152
a' 0
I (\
20
30
0
40"' z
"' 1-.., ~a::
1-z .., 0
60 a:: .., Cl.
70
80
90
100
~
SANO GRAVEL t-~ ~.F~IN:-;=E~~-,-=...c.c..=..~M~E~D~,~u~M,-~~c=o~A~R=s=E=-+-~---:F=1=N=E_.,,:!..!.'.:~=-=c~o~A~R~S~E~~coeeLES
GRADATION TEST
ANALYSIS CURVE OF MATERIAL USED FOR STUDY
,, G> C :::0 rr,
TEMPLATE USED FOR SHAPING CHANNEL
/-~--,,\ '2" D.
I 1 ---- 3 I ----1-f ,-,)i._--- - -1.\ - -+--
--l-- +-\ -~_;I . / \ I \ I
\ / ' . / ,, .,.../ ...... __ --
PLAN
_!..!FL=-"O'-"W'--->-
f-c-------------- 3"!-------- --.~----: =L()
: I" : k -------1!-------1 o
-f:, _J ' i . T ' i ':9. ·-ia) Vo
' ' ' ' '
' '
' "-IC\J a,
' '
1"
--. I.D.
---BRACE
--CONNECT TO WELL
ELEVATION
STATIC DISK
STILLING WELL WITH HOOK GAGE-- --------
ELEVATION
FIGURE 6
,-MOVEABLE / ATTACHMENT I
~ T J I \ T : \ __ :( __ -+--+----+-t<H
I I
I I I I I I I
I I
I
I I I
"-IN "'
I I I I I I
I I I
I "-IN
---\ _ _I __
-,'-SEE SECTION OF CL IPS
I I I I
I
,7SOLDER
I I
I I I I I I __ y _______________ r--1--+~----,f+t-(
I 19111 I I I I I I I I I
'-IN I'-
0
-- -BRASS WEB
//PITOT STATIC TUBE /
!---
PITOT BANKS 6 7
SCALE OF INCHES
F-------- 11 I Iii I
I I I I I I I
I I I I I I I I I I
I I I I I I I I I I
I I
I
I I
I I I
+----- -- 111 i I I I I I I I I I I I I I I I
l ;_]N -T--
_L __ ~T __ _
FIGURE 7
(a) Pitot Tube and Cylinder Bank
Figure 8
(b) Mounted Pitot Banks
(c) Sloping Manometer Board
PITOT TUBES AS OPERATED
(/)
z 0
I-<t: > w _J
w
w > -I-<t: _J
w a::
6.0
5 .8
5.6
5.4
5.2
6.0
5.8
5.6
5.4
5.2
6.0
5.8
5.6
5.4
5.2
6.0
5.8
5.6
5.4
5.2
5.0
' ' ' '
2 I 0 I 2 DISTANCE LEFT OF i-FEET
I DISTANCE RIGHT OF i-FEET
......................
'\"--\
\ \
\
' ' ' ', .... _:::: -TEST 8
I
I ---------------
STATION 0+10
I I
I <t.
----------------STATION 0+30
TEST
-
/ /
I I
I
1---::. .... ........ ,,
/ /
/ /
/
7 /
/ /
/ /
/ /
-7 /
/
,/ .....-r
/ I / I
_..- I
- I ~ I
------- -----STATION O+ 50
/ /
I I
TEST I-<~--
--------- ---------STATION 0+72
CROSS SECTION CHANGES DURING TESTS
I I
I I
:c: l
FIGURE 10
a.. 0.3 i--.~-~---!----~.___--- ---l----#---+--#----:a~'l--------1 w C)
........ >a:: <{ 0 0 .2 l--------!----,~----+--------:,~-~~-----:,~C,----+------1 z => 0 CD
~ 0 a:: u..
--1' - o" I
----s" I
- - - - - 4"
w (.) z O .I lc-----------4--~.___-----loll':....l----"-"~-+------+------ -+-------1
<{ 1-(/) 0 .08 1---------!~.___~'-l---+l-l-l-------1--------'-------'--------1 C)
II NOTE : C)
........ >- 0.06 l-----~--,j~'---~'-l----+------1
NUMBERS REFER TO DISTANCE RIGHT OF£
DISTANCE FROM BOUNDARY IS ME ASU RED AT RIGHT ANGLES TO BOUNDARY
0 .03 .--------!------+---- ---1------+-------t--------i
0 .02 L--___ ____. ___ _ --1. ___ _ __._ ___ _ ....._ ___________ _
1.0 1. 2 1.4 1.6 1.8 2 .0 2.2 VE LOCI TY - FEET /SECOND
VE LOCI TY PROF! LES FOR TEST 5
N I 0 xa~---~-~------~-------~------~-~
II "" . I-
LL ..........
/---TEST 8 '
~6~---~-~~~~---~---~~ ~-~------~-~ ..J
z
~ TEST ~4~---~--~~~~--~~- ~----~------~-~ 0 LL
w >
TEST 2----- --/
~2~---t~~----~~~~~~~~~~~=~~~~~J:d ~ I-
TEST 1----/ ' ,
o----~--------..___ _______ ___._ ______ __._ _ ___, 0 10 30 50 66 72
DISTANCE IN FT. FROM HEADBOX
TRACTIVE FORCE VARIATION ALONG CHANNEL
"'Tl
G')
C ::0 rn
,,
.
FIGURE
w.s. - --
I
I AVERAGE CROSS SECTION - -- ----.
ct
2.5
(~>- 5 .0 I 7.5 I I TEST I I I I ---- W.S. I -- -I I I I I AVERAGE CROSS SECTION--- --I I
t. I I I I I I
I I 2.5 C\I
i--: LL 5.0 ' (D 7.5 ...J TEST 2 I
C\I w.s. 0 ---X
I
w I AVERAGE CROSS SECTION-- ---0 0:: ct 0 LL
w > ~ 2.5 0 <{ 5.0 0:: I- 7.5
TEST 3
W.S. -- -
I
I AVERAGE CROSS SECTION------
ct
I ' s- 5.0 \----------i-----------1\-----------1\----- ----1
7.5 t--------t----- -----,t----------,t--------1
0 2 3 DISTANCE RIGHT OF <t,-FEET
TRACTIVE FORCE DISTRIBUTIONS TESTS I THROUGH 4
TEST 4
12
I
I <t
FIGURE 13
w.s.
AVERAGE CROSS-SECTION--.,
,->- 5.0 ~---=~~=~1::'.::::::::::'.:_ ___ ___j~-~c_ __ ___j ____ __j I
J 7.5 1-------------l-----------l-----------l------, 1 TEST 5 I I I
I I I I I I
C\J ILL.. ......... CD ...J I
C\J 0
X
w L) a:: e
w.s .
i AVERAGE CROSS-SECTION --•• ,
<t
2.5 ~~~~~ --+-~~~ --+--:;;;;;,,,"'~::::~::::::::~~:::.___~~~ 5 .0 ~~~-::;;;0~~:::::~::::::::::::____--+-~~~~~-l---~~--l 7.5 1---------1------------'l-------------l------,
TEST 6
w.s.
i AVERAGE GROSS- SECTION-.,
',
Cr.
w > i'.= 2.5 1-------1---- -+-----Jb.6---~=----l------l L) <l: 5.0 .1:-------+--~---L---+-------+------, ~ 1.5 ~=~==~"""""-~L=--------1--------+---~
TEST 7
I
I I ft I I I I I
w.s.
AVERAGE CROSS-SECTION -.\
'
I 2.5 f---------1--- ----1--+-+---~-"""°'~+---------l I
'->- 5.0 ........,,------+------1----++--,,,,,,::.-----+--------l
0 2 3
DISTANCE RIGHT OF <i. , PEET
TRACTIVE FORCE DISTRIBUTIONS TESTS 5 TH ROUGH 8
.. •
2 .0 ..--------.------.----------.-• -------,------.-----,-------,-------r------------, ®
• 0 X
El • 2 El ____ Tr = 1.269 ++.05_9Dr - 2.312Dr
1.5 1-------t------:«,,----+-----""7"1------t-----+-------+-------l
• 0
X
•
®
X •
0
X
TRACTIVE FORCE ACTING AT BOUNDARY POINT
AVERA GE TRACTIVE FORCE
X
El
•
00 X
•
0
®
•
. -X
El 0
IKI
®. -0-
0
KEY
TEST I TEST 2 TEST 3 TEST 4 TEST 5 TEST 6 TEST 7 TEST 8
0'-----~----...._ ____ .__ ___ ~ ____ ...._ ____ .__ ___ ~ ____ _._ ____ ..__ ___ _
0 0.1 0.2 0 .3 0 .4 0.5 0 .6 0.7 0 .8 0. 9 1.0
DISTANCE RIGHT OF CENTER LINE
DISTANCE TO WATER SURFACE EDGE Dr
T RACTIVE FORCE DISTRIBUTION
,, Ci)
C ::u rn
'
•
10
9
en 8 LL g
7 w (!)
~ <( 6 I <.) CJ)
05
4
19
.., I 018
X
II
=c 17
CJ)
(!) zl6
z z <C 15 ~
14 0
FIGURE 15
--D.~ .-,08
7 ~------FROM T=YrSe---, ~-6 ):,- -;:-'
0/ ·' 5 ~D.A, __ -FROM 2
4 0-~ ~ - T= C(Uy2-UY1l .... -
3 D."'f:,-~
y 2b'J
/.l 1J D./
I 2 3 4 5 AVERAGE TRACTIVE FORCE AT STA. 0+66=X I0-2
(a)
TEST I 2 3 4 5 6 7 8
MANN IN G'S 0.0144 0 .0144 0.0145 0 .0150 0.0153 0.0156 0.0167 0.0180 "n'' VALUE
/ /08
,/ Ii'"
~ / 07
5 ,~ ...
0/ ~0- 06.....- - -TEST NUMBERS
.,,,. 4
0-0-0" ~ I 2 3
I 2 3 4 5 AVERAGE TRACTIVE FORCE AT STA. O+ 66 = X 10-2
T = Yr Se
( b)
COMPARISON OF THE AVERAGE TRACTIVE FORCE FROM THE TWO METHODS
AND INCREASE IN MANNING1
S 11 n11 VALUE
6
6
GPO 845387
\ ..
•
)